Second Order Differential Equations: An Introduction

Stretch a spring and let go. What happens next?

You could try to describe it in words: it snaps back, overshoots, comes back again, a little less each time, until it settles. But there's a tighter way to say all of that in one line, a single equation that already knows the whole story before it unfolds. That's what a second order differential equation is: a rule that ties together where something is, how fast it's moving, and how fast that speed itself is changing, all at once.

Learn to read that rule, and you can predict a spring, a swing, a plucked guitar string, and a bridge in the wind, all from the same idea.

What does "second order" even mean?

A first order equation talks about a rate of change: how fast a quantity moves right now. That's velocity, the derivative of position. Simple enough, your speedometer shows it to you directly.

A second order equation goes one level deeper. It talks about how the rate of change is changing, that's acceleration, the derivative of velocity. Press the gas pedal and your speed doesn't jump, it climbs, and how quickly it climbs is exactly what a second order equation describes.

We write position as a function of time, $x(t)$. Its first derivative $x'(t)$ is velocity. Its second derivative $x''(t)$ is acceleration. A second order differential equation is simply an equation that includes $x''$. And once acceleration is in the picture, so is Newton's second law, $F = ma$, which is exactly why these equations show up everywhere in physics.

The simplest example: a mass on a spring

Here's the whole setup. Hang a mass from a spring. Pull it down a bit, then let go. The spring pulls back with a force proportional to how far it's stretched, and always pointing back toward the resting position. That's Hooke's law:

where $x$ is the displacement from rest and $k$ measures the spring's stiffness. Now combine that with Newton's second law, $F = ma = mx''$. Set the two forces equal, and you get:

Read that equation out loud: acceleration is proportional to position, but pointing the opposite way. The farther the mass gets pulled from rest, the harder it gets yanked back. That's the whole mechanism behind every oscillation you've ever seen.

Slide stiffness up and the mass snaps back and forth faster, the frequency climbs with it. Notice the period shrinks exactly like $T = 2\pi\sqrt{m/k}$ predicts, stiffer spring, faster wiggle, every single time.

What happens when there's friction?

Real springs don't bounce forever. Air resistance, internal friction, something always drains the motion, and that drag force is roughly proportional to speed, always opposing whichever way the mass is currently moving:

The new term, $cx'$, is friction. Divide through by $m$ and give the two remaining pieces names, $\omega = \sqrt{k/m}$ for the natural frequency and $\zeta = c/(2\sqrt{km})$ for the damping ratio, and the equation simplifies to:

Now here's the trick for solving it. Guess that the solution looks like $x(t) = e^{rt}$, some kind of exponential, and plug it in. Every derivative of $e^{rt}$ is just another copy of itself times $r$, so the whole equation collapses down to a plain algebra problem, the characteristic equation:

Solve for $r$, and everything about the motion is hiding in whether those roots come out real or complex.

Push $\zeta$ up from zero and watch the two dots in the left panel. They start as a complex pair, mirrored above and below the real axis, and that's exactly when the curve on the right oscillates, a decaying wiggle inside a shrinking envelope. Keep pushing, and the dots slide onto the real axis and split apart into two ordinary real numbers... and the wiggle vanishes entirely. The mass just glides back to rest, no overshoot. Right at the crossover sits the fastest possible non-oscillating return, called critically damped, which is exactly what a good car's shock absorbers are tuned for.

Why do we need two starting facts, not one?

Here's a question worth sitting with: if you know the equation, do you know the future? Not quite, not yet. Two springs governed by the identical equation can behave completely differently, one barely stretched and released gently, another yanked far out and shoved hard on release.

The missing piece is the state of the system at the start, and that state has two parts: where it is, $x(0)$, and how fast it's moving, $x'(0)$. Knowing position alone doesn't tell you which way things are headed next. You genuinely need both.

Drag the dot anywhere on that grid, it's setting position on the horizontal axis and velocity on the vertical one, together in one move. The little arrows show which way the state drifts from every point, that's the rule itself, drawn as a flow. Release the dot and it slides along that flow, tracing out the entire future in one shot. Same equation, different starting state, completely different path. That's why a second order equation always needs exactly two initial conditions to pin down one unique answer.

What if something keeps pushing?

So far the spring has been left alone after the initial pull. But real systems often get pushed continuously, a swing pumped rhythmically, an engine's vibration rattling a wing, a radio tuner picking one station out of the air. Add a repeating push, $F_0 \cos(\Omega t)$, to the equation and something dramatic can happen depending on how that push's rhythm, $\Omega$, compares to the system's own natural rhythm, $\omega$.

Push at the wrong rhythm and the pushes partly fight the motion, partly help it, mostly cancelling out into a modest wobble. But push at nearly the same rhythm as $\omega$, and every push lands at exactly the right moment to add more energy in... and the swings grow enormous.

Slide the drive frequency toward $\Omega/\omega = 1$ and watch the amplitude curve spike, the mass on the right starts swinging wildly from a tiny push. This is resonance, and it's the same effect that lets a trained voice shatter a wine glass and the same effect engineers spend enormous effort designing bridges to avoid.

The short version

A second order differential equation ties position, velocity, and acceleration together in one rule, and because acceleration is what forces produce, these equations are just Newton's second law written for whatever is pushing and pulling on something. A plain spring oscillates forever. Add friction and it settles down, smoothly or with a dying wiggle, depending on how strong that friction is relative to the stiffness. Two numbers, starting position and starting velocity, are always enough to pin down the entire future. And if you push the system at its own natural rhythm, small pushes turn into huge swings, which is either a disaster or a radio tuner, depending on whether you meant for it to happen.

A spring bouncing on your desk and a skyscraper swaying in an earthquake are, underneath, the same three lines of algebra.

All visualizations are interactive React components running entirely in your browser, using SVG. No libraries beyond React.